course overview and introduction
play

Course Overview and Introduction Probabilistic Graphical Models - PowerPoint PPT Presentation

Course Overview and Introduction Probabilistic Graphical Models Sharif University of Technology Soleymani Spring 2017 Course info Instructor: Mahdieh Soleymani Email: soleymani@sharif.edu Website:


  1. Course Overview and Introduction Probabilistic Graphical Models Sharif University of Technology Soleymani Spring 2017

  2. Course info  Instructor: Mahdieh Soleymani  Email: soleymani@sharif.edu  Website: http://ce.sharif.edu/cources/95-96/2/ce768-1  Lectures: Sun-Tue (13:30-15:00) , Room 202  Teacher assistants:  Maryam Ramezani  Amirhossein Akbarnejad  Seyed Ali Osia  Seyed Mohammad Chavosian 2

  3. Text book  D. Koller and N. Friedman, “ Probabilistic Graphical Models: Principles and Techniques ” , MIT Press, 2009.  M.I. Jordan, “ An Introduction to Probabilistic Graphical Models ” , Preprint .  Other C.M. Bishop, “ Pattern Recognition and Machine Learning ” , Springer, 2006.  Chapters 8-11,13.  K.P. Murphy, “ Machine Learning:A Probabilistic Perspective ” , MIT Press, 2012.  3

  4. Evaluation policy  Mid-term: 25%  Final: 35%  Home works & course works: 40% 4

  5. Why using probabilistic models?  Partial knowledge of the state of the world  Noisy or incomplete observations  We may not know or cover all the involved phenomena in our model  Partial knowledge can cause the world seems to be stochastic  To deal with partial knowledge and/or stochastic worlds we need reasoning under uncertainty 5

  6. Why using a graph for representation?  Intuitively appealing interface by which we can models highly interacting sets of variables  It allows us to design efficient general purpose inference algorithms 6

  7. Probability review  Marginal probabilities  𝑄 𝑌 = 𝑧 𝑄(𝑌, 𝑍 = 𝑧)  Conditional probabilities 𝑄(𝑌,𝑍)  𝑄 𝑌|𝑍 = 𝑄(𝑍)  Bayes rule: 𝑄 𝑍|𝑌 𝑄(𝑌)  𝑄 𝑌|𝑍 = 𝑄(𝑍)  Chain rule: 𝑜  𝑄 𝑌 1 , … , 𝑌 𝑜 = 𝑗=1 𝑄(𝑌 𝑗 |𝑌 1 , … , 𝑌 𝑗−1 ) 7

  8. Medical diagnosis example  Representation 𝑒 1 𝑒 2 𝑒 3 𝑒 4 diseases Findings 𝑔 𝑔 𝑔 𝑔 𝑔 5 1 2 3 4 (symptoms & tests) 𝑄(𝑔 1 |𝑒 1 ) 𝑄(𝑔 2 |𝑒 1 , 𝑒 2 , 𝑒 3 ) 𝑄(𝑔 3 |𝑒 3 ) … 8

  9. PGMs as a framework  General-purpose framework for representing uncertain knowledge and learning and inference in uncertain conditions .  A graph-based representation as the basis of encoding a complex distribution compactly  allows declarative representation (with clear semantics) of the probabilistic knowledge 9

  10. PGMs as a framework  Intuitive & compact data structure for representation  Efficient reasoning using general-purpose algorithms  Sparse parameterization (enables us to elicit or learn from data) 10

  11. PGM: declarative representation  Separation of knowledge and reasoning  We need to specify our model for a specific application that represents our probabilistic knowledge  There is a general suite of reasoning algorithms that can be used. 11

  12. History  Wright 1921, 1934 and before  Bayesian networks are independently developed by Spiegelhalter and Lauritzen in statistics and Pearl in computer science in the late 1980 ’ s  First applications (1990 ’ s): expert systems and information retrieval 12

  13. PGMs: some application areas  Machine Learning and computational statistics  Computer vision: e.g., segmenting and denoising images  Robotics: e.g., robot localization and mapping  Natural Language Processing  Speech recognition  Information Retrieval  AI: game playing, planning  Computational Biology  Networks: decoding messages (sent over a noisy channel)  Medical diagnosis and prognosis  … 13

  14. Graphical models: directed & undirected  Two kinds of graphical models:  Directed: Bayesian Networks (BNs)  Undirected: Markov Random Fields (MRFs) B A A B C C D D Causality relations Correlation of variables 14

  15. Graphical models: directed & undirected [Pathfinder Project, 1992] 15

  16. Representation, inference, and learning  We will cover three aspects of the graphical models:  Representation of probabilistic knowledge  Inference algorithms on these models  Using the data to acquire the distribution 16

  17. Representation, inference, and learning  Representation: When variables tends to interact directly with few other variables (local structure)  Inference: answering queries using the model  algorithms for answering questions/queries according to the model and/or based given observation.  Learning of both the parameters and the structure of the graphical models 17

  18. Medical diagnosis example  Representation 𝑒 1 𝑒 2 𝑒 3 𝑒 4 diseases Findings 𝑔 𝑔 𝑔 𝑔 𝑔 5 1 2 3 4 (symptoms & tests)  Inference: Given symptoms, what disease is likely?  Eliciting or learning the required probabilities from the data 18

  19. Representation: summary of advantages  Representing large multivariate distributions directly and exhaustively is hopeless:  The number of parameters is exponential in the number of random variables  Inference can be exponential in the number of variables  PGM representation  Compact representation of the joint distribution  Transparent  We can combine expert knowledge and accumulated data to learn the model  Effective for inference and learning 19

  20. Graph structure  Denotes conditional dependence structure between random variables  One view: Graph represents a set of independencies  Another view: Graph shows a skeleton for factorizing a joint distribution 20

  21. Image denoising example 21 [Bishop]

  22. Genetic pedigree example A0 B0 A B Bg Ag A1 B1 D C C0 D0 Cg Dg C1 D1 E E0 Eg E1 22

  23. Plan in our course  Fundamentals of Graphical Models:  Representation  Bayesian Network  Markov Random Fields  Exact inference  Basics of learning  Case studies: Popular graphical models  Multivariate Gaussian Models  FA, PPCA  HMM, CRF, Kalman filter  Approximate inference  Variational methods  Monte Carlo algorithms 23

Recommend


More recommend